Advanced

Toolbox for development and validation of grey-box building models for forecasting and control

De Coninck, Roel; Magnusson, Fredrik LU ; Åkesson, Johan and Helsen, Lieve (2016) In Journal of Building Performance Simulation, Taylor & Francis 9(3). p.288-303
Abstract
As automatic sensing and information and communication technology get cheaper, building monitoring data becomes easier to obtain. The availability of data leads to new opportunities in the context of energy efficiency in buildings. This paper describes the development and validation of a data-driven grey-box modelling toolbox for buildings. The Python toolbox is based on a Modelica library with thermal building and Heating, Ventilation and Air-Conditioning models and the optimization framework in JModelica.org. The toolchain facilitates and automates the different steps in the system identification procedure, like data handling, model selection, parameter estimation and validation. To validate the methodology, different grey-box models are... (More)
As automatic sensing and information and communication technology get cheaper, building monitoring data becomes easier to obtain. The availability of data leads to new opportunities in the context of energy efficiency in buildings. This paper describes the development and validation of a data-driven grey-box modelling toolbox for buildings. The Python toolbox is based on a Modelica library with thermal building and Heating, Ventilation and Air-Conditioning models and the optimization framework in JModelica.org. The toolchain facilitates and automates the different steps in the system identification procedure, like data handling, model selection, parameter estimation and validation. To validate the methodology, different grey-box models are identified for a single-family dwelling with detailed monitoring data from two experiments. Validated models for forecasting and control can be identified. However, in one experiment the model performance is reduced, likely due to a poor information content in the identification data set. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Journal of Building Performance Simulation, Taylor & Francis
volume
9
issue
3
pages
288 - 303
publisher
Taylor & Francis
external identifiers
  • scopus:84959112290
  • wos:000374992500004
ISSN
1940-1507
DOI
10.1080/19401493.2015.1046933
project
collocation
LCCC
language
English
LU publication?
yes
id
f4995b32-00e2-4c69-a890-30848a0402e7 (old id 7366117)
date added to LUP
2015-06-29 10:02:37
date last changed
2017-06-25 03:24:19
@article{f4995b32-00e2-4c69-a890-30848a0402e7,
  abstract     = {As automatic sensing and information and communication technology get cheaper, building monitoring data becomes easier to obtain. The availability of data leads to new opportunities in the context of energy efficiency in buildings. This paper describes the development and validation of a data-driven grey-box modelling toolbox for buildings. The Python toolbox is based on a Modelica library with thermal building and Heating, Ventilation and Air-Conditioning models and the optimization framework in JModelica.org. The toolchain facilitates and automates the different steps in the system identification procedure, like data handling, model selection, parameter estimation and validation. To validate the methodology, different grey-box models are identified for a single-family dwelling with detailed monitoring data from two experiments. Validated models for forecasting and control can be identified. However, in one experiment the model performance is reduced, likely due to a poor information content in the identification data set.},
  author       = {De Coninck, Roel and Magnusson, Fredrik and Åkesson, Johan and Helsen, Lieve},
  issn         = {1940-1507},
  language     = {eng},
  number       = {3},
  pages        = {288--303},
  publisher    = {Taylor & Francis},
  series       = {Journal of Building Performance Simulation, Taylor & Francis},
  title        = {Toolbox for development and validation of grey-box building models for forecasting and control},
  url          = {http://dx.doi.org/10.1080/19401493.2015.1046933},
  volume       = {9},
  year         = {2016},
}